A fire environment emergency evacuation system for a railway tunnel emergency rescue station

Abstract

The present invention provides a fire environment emergency evacuation system for a railway tunnel emergency rescue station, which includes: a subsystem and a central system; the subsystem includes: a data acquisition module for real-time monitoring of various disaster-causing factor data in the area where the subsystem is located; a data analysis module for using a parallel computing judgment algorithm with time filtering to determine the danger level of the area by threshold comparison; an alarm module for issuing an alarm and indicating the safe evacuation direction based on the danger level; and a central system for aggregating data from all subsystems, dynamically dividing safe evacuation areas on a preset two-dimensional grid map, and calculating the optimal evacuation path from each grid to the safe area based on an improved D* algorithm that takes into account the special environment of railway tunnel fires to avoid the dangerous area, ultimately obtaining the optimal evacuation path that takes into account the real-time changes in the fire environment. This embodiment can significantly improve the safety assurance capability of railway tunnels and is of great significance to promoting the sustainable and healthy development of railway transportation.

Description

Technical Field

[0001] The present invention relates to the field of traffic safety, and in particular to a fire environment emergency evacuation system for a railway tunnel emergency rescue station. Background Art

[0002] Currently, both domestic and international research has achieved considerable success in tunnel fire monitoring and early warning technologies. For example, some research has developed tunnel fire prevention technologies based on various types of detectors, which can monitor fires in real time and automatically trigger alarms and activate firefighting equipment when they occur. Others have developed tunnel environmental monitoring systems that automatically activate firefighting and ventilation systems after a fire occurs, ensuring air quality within the tunnel. However, existing research is limited to monitoring fire activity and assessing the overall fire development. Monitoring the deterioration of the evacuation environment during a fire has not been achieved. In particular, there is little research on real-time monitoring and early warning of hazard factors in the evacuation environment during a fire at a railway tunnel emergency rescue station, as well as systematic dynamic evacuation strategy control. In other words, existing monitoring and early warning devices or systems are unable to monitor and warn of hazard factors in the tunnel evacuation environment during a fire, nor can they output the optimal evacuation route in real time. Summary of the Invention

[0003] In order to solve the problems existing in the prior art, the present invention provides a fire environment emergency evacuation system for a railway tunnel emergency rescue station to improve the above problems.

[0004] The present invention is achieved in that:

[0005] A fire environment emergency evacuation system for a railway tunnel emergency rescue station, comprising:

[0006] Subsystems and a central system; the subsystems are installed on the side walls of the emergency rescue station at a certain interval, and the data collected by the subsystems are transmitted to the central system via serial communication; wherein:

[0007] The subsystem includes:

[0008] Data collection module, real-time monitoring of various disaster-causing factors in the area where the subsystem is located;

[0009] The data analysis module uses a parallel computing algorithm with time filtering based on various disaster-causing factor data to determine the danger level of the area through threshold comparison.

[0010] An alarm module, which issues an alarm according to the degree of danger and indicates the safe evacuation direction;

[0011] The central system aggregates data from all subsystems, dynamically divides safe evacuation areas on a preset two-dimensional grid map, and calculates the optimal evacuation path from each grid to the safe area based on an improved D* algorithm that takes into account the special environment of railway tunnel fires to avoid dangerous areas. Ultimately, the optimal evacuation path is obtained that takes into account real-time changes in the fire environment, and then the signal is transmitted back to each subsystem to indicate the evacuation direction.

[0012] Preferably, the data acquisition module includes a temperature sensor, an oxygen concentration sensor, a thermal radiation sensor, a CO concentration sensor, and a visibility sensor; each sensor transmits the collected data to the data analysis module through analog signal transmission and digital signal transmission.

[0013] Preferably, the data analysis module includes a sensor interface, a microprocessor unit and a communication interface; wherein:

[0014] The sensor interface is used to receive different types of environmental parameters transmitted by various sensors;

[0015] The microprocessor unit is configured to receive the environmental parameter data transmitted by the sensor interface and preprocess the environmental parameters to convert them into a unified format that can be processed by the microprocessor unit; wherein, during the preprocessing, a timer is defined to match and align the environmental parameters of different sensors according to timestamps at a preset standard time interval to ensure data consistency and accuracy, and then the environmental parameters and corresponding timestamps are respectively filled into corresponding structure fields to obtain the environmental parameters in a unified format;

[0016] The microprocessor unit is also used to compare the environmental parameters processed by different sensors with preset thresholds to determine whether the environmental parameters exceed their corresponding thresholds. Based on the situation of exceeding the thresholds, a time-filtered parallel computing environmental parameter safety judgment algorithm is used to make a safety judgment on the environmental parameters. When the area where the judgment subsystem is located is determined to be a dangerous area, an alarm signal is sent to the alarm module through the communication interface.

[0017] Preferably, the specific judgment process of the time-filtered parallel computing environment parameter security judgment algorithm is as follows:

[0018] Define the threshold judgment function Fi (i=1, 2, ..., N) of N environmental parameters;

[0019] The data Pi (i=1, 2, ..., N) collected in real time from N environmental parameters are simultaneously used as input and passed into each judgment function Fi for calculation;

[0020] The judgment function Fi calculates whether the corresponding data Pi exceeds its corresponding threshold Ti in a parallel computing environment and records the result as Ri; where Ri is 0 or 1, 0 means that the threshold is not exceeded, and 1 means that the threshold is exceeded;

[0021] Perform time filtering on the N results Ri; if Ri is 1 in consecutive ΔT time slices, the environmental parameter is finally judged to be above the threshold and the area is judged to be a dangerous area; otherwise, it is a safe area.

[0022] Preferably, the subsystem also includes a test / reset button, a backup power supply and a status indicator; wherein the test / reset button provides a physical button for testing the alarm function or resetting the alarm status, and the status indicator uses an LED indicator light to display the working status of the module; the backup power supply is used to ensure that the subsystem itself has a stable power supply.

[0023] Preferably, the central system is the part that connects the subsystems and performs data processing, evacuation path analysis, and sends feedback signals. The central system uniformly processes the collected data, and dynamically divides the safe evacuation area in a preset two-dimensional grid map. It also calculates the optimal evacuation path from each grid to the safe area based on the improved D* algorithm combined with the special environment of railway tunnel fires, thereby avoiding the dangerous area. It can obtain a real-time optimal evacuation path according to changes in the evacuation environment, and finally transmits the signal back to each subsystem, indicating the safe evacuation direction according to the direction of the evacuation path.

[0024] Preferably, the central system includes a data receiving and pre-processing module, a data storage module and an optimal evacuation path analysis module; wherein:

[0025] The data receiving and preprocessing module is used to receive raw data from various subsystems or external data sources, and perform preliminary cleaning, formatting and standardization on the data to ensure data consistency and availability. After processing, the worker thread writes the preprocessed data into a temporary buffer. The main thread periodically extracts data from the temporary buffer and writes it in batches into the data storage module.

[0026] The data storage module is used to store the pre-processed data in an appropriate storage system so that the optimal evacuation path analysis module can access it in real time and efficiently.

[0027] The optimal evacuation path analysis module is used to build a dynamic two-dimensional grid map based on the real-time changing parameters of the tunnel evacuation environment. It then calculates the shortest path from each grid to the safe area based on the improved D* algorithm, corresponding to the optimal safe evacuation path for people in different areas. The signal is then transmitted back to each subsystem based on the optimal evacuation path.

[0028] Preferably, the optimal evacuation path analysis module is specifically used to:

[0029] The optimal evacuation path is calculated using the A* algorithm, and the h(n) and g(n) values ​​for each grid cell are recorded. The h(n) value represents the heuristic function-estimated distance from the grid cell to the target point, which is the entrance to each cross-channel. The g(n) value represents the sum of the actual movement costs from the starting point to the grid cell. The h(n) value for each grid cell is initialized to the Manhattan distance to each cross-channel entrance. g(n) is initialized to infinity, indicating that no exploration has taken place. All cross-channel entrances are considered as candidate destinations.

[0030] Specifically:

[0031] Use the A* algorithm to perform a breadth-first search from the starting point to calculate the shortest path from the starting point to each entrance;

[0032] Every time a grid is explored, its g(n) value is updated to the sum of the actual costs from the starting point to this grid;

[0033] Record all optimal evacuation route plans;

[0034] When the state of a certain grid is detected to be dangerous, its h(n) value is set to infinity, that is, the grid will not be included in the optimal evacuation path;

[0035] With this grid as the center, add the 8 adjacent grids in the direction to the open list;

[0036] Take a grid from the open list for modification, calculate the estimated cost g(n)+h(n) of walking from its adjacent grid to this grid, and compare it with the current g(n) value;

[0037] If the new value is better, update g(n) with the new value and add the adjacent grid to the open list;

[0038] Repeat the above steps until the open list is empty, which indicates the global optimal evacuation path. Finally, output the revised optimal evacuation path.

[0039] Preferably, the optimal evacuation path analysis module is further configured to: after determining the danger level of the area where the compartment door is located, when a compartment door is in a dangerous area, generate the optimal evacuation path for the people who have left the compartment using the following algorithm:

[0040] Get the boundary coordinate point set of the danger zone;

[0041] A boundary line is constructed based on all boundary coordinate points in the boundary coordinate point set, and the distance from the starting point to each point on this boundary line is calculated. The point with the minimum distance is selected as the temporary target point candidate on the boundary line.

[0042] Select the candidate temporary target point with the smallest distance as the global temporary target point, and modify the h(n) values ​​of the dangerous area and the global temporary target point;

[0043] Use the D* algorithm to calculate the shortest path out of the danger zone, then restore the h(n) value, restart the D* algorithm, and calculate the complete optimal evacuation path.

[0044] Preferably, the optimal evacuation path analysis module is further used to:

[0045] After determining the danger level of the area where the car door is located, if a car door is in the danger zone, personnel who have not left the car will be guided or warned by emergency stations and in-car broadcasts to guide people from the current car to other cars and then to the emergency station. The following algorithm is used to generate the optimal evacuation path:

[0046] The area outside the carriage is designated as a no-traffic zone, and people are no longer evacuated through this area;

[0047] Instead of setting the car door as the starting point, the entrance to the internal passage of the safety car connected to the car is set as the new starting point. At the same time, determine whether the car door adjacent to the danger zone needs to be temporarily blocked to prevent people from accidentally entering the danger zone. If blocking is required, remove it from the passable area and modify the h(n) values ​​of the danger zone and the target point.

[0048] Recalculate the optimal evacuation path from the new starting point, that is, the entrance of the internal passage of the adjacent safety compartment through the internal connecting passage of the compartment to the entrance of each horizontal passage.

[0049] This embodiment can capture early fire signals in real time, collect and analyze disaster-causing factors within emergency rescue stations in real time, issue alarm signals within dangerous areas, and, based on the impact of disaster-causing factors in railway tunnel fires on personnel evacuation behavior, plan optimal evacuation routes in real time and direct personnel evacuation directions in fire situations. This embodiment not only protects the lives of passengers and staff and reduces property losses in the event of a fire accident, but also improves the safety management level of railway tunnels and ensures the safety of railway transportation. The implementation and application of this embodiment is expected to significantly enhance the safety assurance capabilities of railway tunnels and is of great significance to promoting the sustainable and healthy development of railway transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0051] Figure 1 A schematic structural diagram of a fire environment emergency evacuation system for a railway tunnel emergency rescue station provided by an embodiment of the present invention;

[0052] Figure 2 It is a structural diagram of the subsystem;

[0053] Figure 3 It is a structural diagram of the data acquisition module;

[0054] Figure 4 It is a structural diagram of the data analysis module;

[0055] Figure 5 It is a structural diagram of the alarm module;

[0056] Figure 6 This is the working principle diagram of the optimal evacuation path analysis module. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0058] See also Figure 1 The embodiment of the present invention provides a fire environment emergency evacuation system for a railway tunnel emergency rescue station, which includes:

[0059] Subsystem 20 and central system 10; the subsystem 20 is installed on the side wall of the emergency rescue station at a certain distance, and the data collected by the subsystem 20 is transmitted to the central system 10 via serial communication.

[0060] In this embodiment, according to the "Design Specification for Disaster Prevention, Evacuation and Rescue Engineering in Railway Tunnels" (TB10020-2017), the spacing between transverse passages within an emergency rescue station should not exceed 60 meters. A passenger train carriage is typically about 25 meters long. Therefore, when the evacuation danger zone is located between two transverse passages, to prevent evacuees within this section from evacuating in the wrong direction of the danger zone, the spacing between subsystems 20 should not exceed 30 meters. In other words, the emergency rescue station platform between two transverse passages should have at least one subsystem 20 located at the beginning, end, and middle of the platform.

[0061] In this embodiment, reference Figure 2 The subsystem 20 includes a data acquisition module 201 , a data analysis module 202 and an alarm module 203 .

[0062] In this embodiment, the data acquisition module 201 monitors the changes in data of various disaster-causing factors in a fire environment in real time; the data analysis module 202 quickly processes the parameters of various disaster-causing factors, compares them with the human body's tolerance threshold, and thus determines the degree of danger; the alarm module 203 activates the sound and light alarm when the environmental parameter data exceeds the threshold; at the same time, the subsystem 20 will also transmit the collected data to the central system 10 through serial communication, so that the central system 10 can perform the next step of optimal evacuation path analysis and signal feedback.

[0063] Specifically, please refer to Figure 3 The data acquisition module 201 includes sensors such as a temperature sensor 2011, an oxygen concentration sensor 2012, a thermal radiation sensor 2013, a CO concentration sensor 2014, and a visibility sensor 2015. Various sensors can collect parameters of different disaster-causing factors in a fire environment in real time and transmit the data to the data analysis module 202.

[0064] Specifically, in this embodiment, temperature sensor 2011 can be the DS18B20, a digital temperature sensor known for its high precision and compact size. It can measure temperatures within a range of -55°C to +125°C with an accuracy of ±0.5°C, enabling the DS18B20 to accurately monitor temperature changes at a fire scene. Oxygen concentration sensor 2012 can be the ME2-O2, a sensor specifically designed to detect oxygen concentration in an environment. It features a fast response time and high sensitivity, effectively monitoring reduced oxygen levels at fire scenes. The thermal radiation sensor 2013 can utilize the MLX90614, a non-contact infrared temperature sensor suitable for monitoring the intensity of thermal radiation from a fire source, helping to assess fire development and spread. The CO concentration sensor 2014 can utilize the Figaro TGS5042, a sensor used to detect carbon monoxide (CO) concentrations. Known for its stability and high sensitivity, it provides accurate CO concentration readings in complex fire environments. The visibility sensor 2015 can utilize the GP2Y1014AU0F, a sensor used to measure particulate matter concentration in the air to assess visibility. Sensors using these examples can effectively monitor critical environmental parameters while maintaining a small footprint, providing accurate data to support evacuation and safety management in emergencies.

[0065] The data analysis module 202 uses a parallel computing algorithm with time filtering based on various disaster-causing factor data to determine the danger level of the area through threshold comparison.

[0066] The alarm module 203 is used to issue an alarm and indicate the safe evacuation direction according to the degree of danger.

[0067] Continue to refer Figure 4 The data analysis module 202 includes a sensor interface 2021, a microprocessor unit 2022, and a communication interface 2023. The sensor interface 2021 is a hardware interface that allows the data analysis module to directly connect to sensors and is used to receive different types of data transmitted by the sensors. In this example, the different types of signals transmitted by the sensors should be converted into a unified format that can be processed by the microprocessor unit. For the digital signal interface, the corresponding interface circuit should be designed on the PCB board and should comply with the standards corresponding to different communication protocols. For example, the DS18B20 temperature sensor should comply with the One-Wire protocol, and the MLX90614 thermal radiation sensor should comply with the I2C protocol. For the analog signal interface, ensure that the analog signal is input to the ADC interface of the microprocessor unit for direct conversion to a digital signal. Different sensors may operate at different voltage levels, so before connecting to the microprocessor unit 2022, level conversion is required to match the operating voltage of the microprocessor unit 2022.

[0068] In this embodiment, the DS18B20 temperature sensor and MLX90614 thermal radiation sensor are directly connected to the microprocessor unit without requiring level conversion. The ME2-02 oxygen concentration sensor requires a boost converter to provide an appropriate supply voltage, and a resistor divider to adjust the voltage range to within the ESP32's ADC input voltage range (0-3.3V). The Figaro TGS5042 CO concentration sensor operates at 5V, so a boost converter is required to provide an appropriate supply voltage. The sensor outputs a current signal, which is converted to a voltage signal through a load resistor and then adjusted to the voltage range through a resistor divider to accommodate the ESP32's ADC input. The Sharp GP2Y1014AU0F visibility sensor outputs an analog voltage signal and operates at 5V, so a level conversion module is required to adapt it to the ESP32's ADC input.

[0069] In this embodiment, before the sensor collects environmental parameter data, a data structure is first defined within the microprocessor unit 2022 to categorize and store the data parameters and their corresponding timestamps. Each parameter corresponds to a field in the structure. For example, temperature corresponds to the Temperature field, thermal radiation corresponds to the Radiation field, carbon monoxide concentration corresponds to the CO field, oxygen concentration corresponds to the Oxygen field, and visibility corresponds to the Visibility field. In addition, a timestamp field should be defined.

[0070] In this embodiment, after the sensors collect different types of environmental parameter data, a timer is defined, with t = 1.0s as the standard time interval. The data from different sensors is matched and aligned according to the timestamps to ensure data consistency and accuracy. The data and corresponding timestamps are then filled into the corresponding structure fields and transmitted to the computing unit within microprocessor unit 2022 via the serial communication interface. In this way, the values ​​of each parameter and their corresponding timestamps can be accessed through the structure fields, eliminating the need for parsing operations, ensuring efficient and real-time data analysis.

[0071] Continue to refer Figure 4 The microprocessor unit 2022 integrates a CPU, memory, and flash memory, and is suitable for control-oriented applications and simple to medium-complexity data processing tasks. In this embodiment, an ESP32 processor can be selected, which is equipped with a dual-core Tensilica LX6 microprocessor with a maximum speed of 240 MHz, providing sufficient computing power to process data and execute complex algorithms. In addition, the processor has abundant memory resources and ports, which can connect and control various sensors. The communication interface 2023 is used to integrate modules that support specific communication protocols so that the data processed by the subsystem can be transmitted to the central system.

[0072] In this embodiment, the threshold values ​​of different environmental parameters are used as pre-stored values ​​in the microprocessor unit 2022 so that the program can compare the input data with the pre-stored values ​​(ie, the threshold values).

[0073] For example, if the temperature data in the pre-stored value is 60°C, if the parameter is greater than 60°C, it is considered to have exceeded the threshold. If the thermal radiation data in the pre-stored value is 4.5kW / ㎡, it is considered to have exceeded the threshold if it is greater than 4.5kW / ㎡. If the CO concentration data in the pre-stored value is 400ppm, it is considered to have exceeded the threshold if it is greater than 400ppm. If the oxygen concentration data in the pre-stored value is 14.0%, it is considered to have exceeded the threshold if it is less than 14%. If the visibility data in the pre-stored value is 10m, it is considered to have exceeded the threshold if it is less than 10m. It should be understood that the pre-stored values ​​can be adjusted according to actual needs, and the present invention will not be elaborated on here.

[0074] In this embodiment, a parallel computing environment parameter safety judgment algorithm with time filtering is further used in the evacuation area danger judgment. This can achieve safety judgment of environmental parameters while considering the impact of data fluctuations, thereby improving judgment accuracy and computing efficiency. The specific judgment steps are as follows:

[0075] Step 1: Define threshold judgment functions Fi (i=1, 2, ..., N) for N environmental parameters. N is the number of environmental parameters. In this embodiment, N is set to 5, including temperature, thermal radiation, carbon monoxide concentration, oxygen concentration, and visibility. It should be understood that the number of N can be modified according to actual needs, such as 3, 4, or 6, and all such solutions are within the scope of the present invention.

[0076] Step 2: The data Pi (i=1, 2, ..., N) collected in real time from N environmental parameters are simultaneously used as input and passed into each judgment function for calculation.

[0077] Step 3: Each judgment function Fi calculates whether the corresponding parameter Pi exceeds its threshold Ti in a parallel computing environment and records the result as Ri. Ri is 0 or 1, 0 means no threshold is exceeded, and 1 means the threshold is exceeded.

[0078] The judgment function form is:

[0079] Fi(Pi) = {

[0080] 0, Pi ≤ Ti

[0081] 1, Pi > Ti}

[0082] Step 4: Perform time filtering on the N results Ri. If Ri is 1 for consecutive ΔT time slices, the environmental parameter is ultimately judged to have exceeded the threshold and the area is considered a dangerous area; otherwise, it is a safe area.

[0083] The ΔT time slice determines the algorithm's tolerance for environmental parameter fluctuations, that is, the system's sensitivity to evacuation environmental parameters. The main impacts of ΔT on the algorithm are as follows: 1) When personnel run to evacuate within the emergency rescue station, they disturb the surrounding smoke, thereby affecting the data values ​​of various fire environmental parameters. Therefore, ΔT can be used in such special environments to change the algorithm's tolerance for environmental parameter fluctuations. Obtaining an appropriate ΔT can balance system accuracy and safety.

[0084] 2) If ΔT is too small, the system will be too sensitive to whether environmental parameters exceed the threshold. This may lead to a false alarm if the environmental parameters in a certain area exceed the threshold for a short period of time due to the movement of people. In this case, the impact on the evacuation of people is relatively small. However, the system's misjudgment will affect the evacuation route planning.

[0085] 3) If ΔT is too large, the system's sensitivity to whether environmental parameters exceed thresholds will be too low. A situation may arise where a parameter in a certain area exceeds the threshold for a long period of time but is not reported as it should be. In this case, the system's sensitivity is too low and the delay is too high, causing evacuees to mistakenly enter dangerous areas and affecting the central system's evacuation route planning.

[0086] 4) In this embodiment, considering the evacuation environment and the speed of the crowd in the tunnel, ΔT is set to 5s.

[0087] Step 5: Repeat steps 1 to 4 at a fixed time interval Δt (set to 1.0 s in this embodiment) to achieve real-time judgment.

[0088] When the area where a certain subsystem 20 is located is determined to be a dangerous area, the data analysis module 202 in the subsystem 20 sends an alarm signal to the alarm module 203 .

[0089] In this embodiment, reference Figure 5 The alarm module 203 includes a serial communication interface 2031 and an audible and visual alarm 2032. The serial communication interface 2031 is used to receive the data processed by the data analysis module. If the data exceeds the threshold, the alarm signal is transmitted to the audible and visual alarm 2032, thereby turning on the alarm.

[0090] In this embodiment, reference Figure 2, the subsystem 20 can also integrate a test / reset button 204, a backup power supply 205, and a status indicator 206. The test / reset button 204 provides a physical button for testing the alarm function or resetting the alarm state. Specifically, corresponding logic is added to the microprocessor unit 2022 software so that when the button is pressed, a cancel alarm signal is issued, all stored data is deleted, the current alarm state is stopped, and the system is restored to a normal monitoring state. This helps to quickly verify the module function during installation or maintenance; the status indicator 206 uses LED indicators to display the module's working status, such as standby, alarm, error, etc., which helps users intuitively understand the system status. Specifically, to implement this function, first, a digital pin is assigned to each LED representing a different state, and then these LED pins are configured as output mode in the initialization part of the program. Secondly, a function is written to control the corresponding LED according to the current state of the system. For example, if the system is in standby mode, the standby LED is illuminated; if the system detects an alarm condition, the alarm LED is illuminated; if a system error occurs, the error LED is illuminated. Finally, in the main loop of the program or at an appropriate location, the aforementioned state control logic is called based on the real-time state of the system. The purpose of the backup power supply 205 is to ensure that the subsystem itself has a stable power supply. Under normal circumstances, the subsystem is powered by a wired power supply. If an emergency occurs, such as a power line interruption, the backup power supply is activated, allowing the subsystem to maintain normal monitoring and alarm functions for a period of time. In the embodiments of the present application, an 18650 lithium-ion battery can be used. However, the nominal voltage of this type of battery is typically 3.7V, and the voltage can reach 4.2V when fully charged. Therefore, in this embodiment, the voltage should be reduced by the MP1584EN switching regulator module to match the operating voltage of the microprocessor unit and the sensor.

[0091] To improve system integration and reliability, this embodiment integrates key components such as the data acquisition module 201, data analysis module 202, status indicator 206, and test / reset button 204 onto a printed circuit board (PCB). This integrated design not only helps optimize space utilization and reduce system complexity, but also improves the stability of electrical connections, thereby enhancing the performance and durability of the entire subsystem. Integrating these core components on the PCB enables a more compact layout, facilitating installation and maintenance, and facilitating future upgrades. Furthermore, the integrated design reduces external wiring, further improving the system's overall reliability and anti-interference capabilities.

[0092] The central system 10 is used to aggregate data from all subsystems 20, dynamically divide safe evacuation areas on a preset two-dimensional grid map, and calculate the optimal evacuation path from each grid to the safe area based on an improved D* algorithm that takes into account the special environment of railway tunnel fires to avoid dangerous areas. Ultimately, the optimal evacuation path is obtained that takes into account real-time changes in the fire environment, and the signal is then transmitted back to each subsystem to indicate the evacuation direction.

[0093] refer to Figure 1 The central system 10 is connected to all subsystems 20 and receives alarm signals and real-time data transmitted from all subsystems 20. The central system 10 mainly consists of a data receiving and pre-processing module 101, a data storage module 102, and an optimal evacuation path analysis module 103.

[0094] In this embodiment, the data reception and preprocessing module 101 is responsible for receiving raw data from various subsystems or external data sources. It performs preliminary data processing, specifically filtering out abnormal data, unifying and standardizing data from all subsystems, and labeling data from different subsystems with information such as acquisition time and location to ensure data consistency and availability, facilitating subsequent data storage and utilization. After processing, the worker thread writes the preprocessed data to a temporary buffer. The main thread periodically extracts data from the buffer and writes it in batches to the data storage module 102, enabling efficient access by the optimal evacuation path analysis module.

[0095] In this embodiment, the optimal evacuation path analysis module 103 is the core of the central system. It is responsible for constructing a dynamic two-dimensional grid map based on the real-time changing parameters of the tunnel evacuation environment. It then calculates the shortest path from each grid cell to a safe area based on the D* algorithm, corresponding to the optimal safe evacuation path for people in different areas. Signals based on the optimal evacuation path are then transmitted back to each subsystem 20. The alarm module indicates the safe evacuation direction based on the evacuation path. In this special environment, evacuation path planning is integrated with dynamic environmental parameters and the algorithm incorporates the opening and closing status of car doors, evacuation of car connecting passages, and the target destinations of multiple transverse passages. This improves the traditional evacuation path planning D* algorithm, enabling safer and more orderly evacuation of personnel and preventing them from entering dangerous areas.

[0096] Furthermore, in specific implementation, the steps include:

[0097] First, based on the structure of train carriages and tunnel emergency rescue station platforms, a two-dimensional grid map of personnel evacuation of corresponding sizes was constructed. Each grid has different status divisions, including safe areas, dangerous areas, end areas (such as the entrances to each horizontal passage), and prohibited areas (such as walls and the edges of carriages).

[0098] Then, during initialization, the area where each car door is located is divided into dangerous areas based on the data signal uploaded by the subsystem. Then, the improved D* algorithm is used, with all train car doors as the starting point and multiple cross-channel entrances as the target end point, combined with the evacuation environment data in the tunnel to perform real-time optimal evacuation path planning.

[0099] The path planning algorithm at this time has two situations. The first is that a certain compartment door is located in a safe area. In this case, it is necessary to classify and discuss it in combination with the danger classification of the surrounding area:

[0100] When there is no dangerous area blocking the carriage door on the route to the cross passage entrance, priority is given to calculating the shortest evacuation path, which is also the optimal evacuation path, with the carriage door as the starting point and the cross passage entrances as the end points; when the adjacent areas of the carriage door are all dangerous areas, and there are no cross passage entrances between the adjacent dangerous areas, then the safe area can be regarded as being "surrounded", and people cannot reach the safe area. At this time, the area should be designated as a dangerous area and is not allowed to be used as a starting point.

[0101] The second situation is that a certain compartment door is originally in or has been changed to a dangerous area. At this time, if there are people who have left the compartment, their primary goal is to leave the dangerous area and then go to the target area. The optimal evacuation path should be recalculated based on this goal. For people who have not left the compartment, the compartment door should be closed to ensure a safe environment inside the car, and then the people inside the car can evacuate to other compartments through the compartment connecting passage.

[0102] Specifically, refer to Figure 6 When a car door is in a safe area and there is no dangerous area blocking the route to the cross passage entrance, the following algorithm is used:

[0103] First, the A* algorithm is used to calculate the optimal evacuation path.

[0104] Among them, the A* algorithm, as one of the heuristic search algorithms, is an algorithm that finds the lowest passing cost for a path with multiple nodes on a graph plane.

[0105] The core of the A* algorithm lies in the design of its evaluation function:

[0106] f(n)=g(n)+h(n)

[0107] Where f(n) is the estimate for each possible trial point, which consists of two parts:

[0108] One part is g(n), and the g(n) value represents the total actual moving cost from the starting point to the grid cell.

[0109] The other part, h(n), represents the heuristic function-estimated distance from the grid cell to the target point, which can be the entrance to different horizontal passages.

[0110] Record the h(n) and g(n) values ​​for each grid cell.

[0111] Use the D* algorithm to correct the optimal evacuation path.

[0112] The D* algorithm is a dynamic path planning algorithm, commonly used to solve path planning problems in dynamic environments. The D* algorithm's name comes from the abbreviation "Dynamic A*," which combines the A* search algorithm with dynamic path planning.

[0113] The evacuation target point at the entrance of the transverse passage is obtained according to the modified optimal evacuation path.

[0114] The specific steps are as follows:

[0115] Step 1: Initialize h(n) of each grid cell in the map to the Manhattan distance to the entrance of each cross channel. The Manhattan distance is the distance between two points in a two-dimensional plane coordinate system. It is the sum of the absolute values ​​of the coordinate differences between the two points. Manhattan distance = |x1-x2| + |y1-y2|; Initialize g(n) to infinity, indicating that it has not been explored.

[0116] Treat all cross passage entrances as alternative endpoints.

[0117] Use the A* algorithm to perform a breadth-first search from the starting point to calculate the shortest path from the starting point to each entrance.

[0118] Every time a grid is explored, its g(n) value is updated to the sum of the actual costs from the starting point to this grid.

[0119] Record all optimal evacuation route plans.

[0120] Step 2: When the state of a certain grid is detected to be dangerous, its h(n) value is set to infinity, that is, the grid will not be included in the optimal evacuation path.

[0121] With this grid as the center, add its 8 adjacent direction grids to the open list.

[0122] Step 3: Take a mesh from the open list to be modified.

[0123] Calculate the estimated cost g(n)+h(n) of walking from its adjacent grid to this grid and compare it with the current g(n) value.

[0124] If the new value is better, update g(n) with the new value and add the adjacent grid to the open list.

[0125] Repeat steps 2 and 3 above until the open list is empty, which indicates the global optimal evacuation path. Finally, output the revised optimal evacuation path.

[0126] After determining the danger level of the area where the carriage door is located, if a carriage door is in the danger zone, the following algorithm is used for people who have left the carriage:

[0127] First, obtain the boundary coordinate point set of the dangerous area;

[0128] Then, a boundary line is formed based on all boundary coordinate points in the boundary coordinate point set, and the distance from the starting point to each point on this boundary line is calculated, and the point with the minimum distance is selected as the temporary target point candidate on the boundary line.

[0129] The danger zone is a portion of the emergency station platform, typically a rectangle. The sidelines are the two short sides of the rectangle. The temporary target point is obtained by calculating the shortest distance from the starting point to the two short sides. The sidelines consist of all boundary coordinate points within the boundary coordinate point set.

[0130] Next, the point with the smallest distance is selected as the global temporary target point, and the h(n) values ​​of the danger zone and the temporary target point are modified.

[0131] Among them, when modifying, the h(n) value of the grid in the danger zone is modified to the heuristic function estimated distance to the temporary target point.

[0132] Finally, the D* algorithm is used to calculate the shortest path out of the danger zone and then the h(n) value is restored. The D* algorithm is restarted to calculate the complete optimal evacuation path.

[0133] If the area where the car door is currently located is originally safe, but becomes a dangerous area as the fire develops, the car door is closed. At this time, for people outside the car who have already left the car, this method can provide the shortest evacuation route.

[0134] After determining the danger level of the area where the car door is located, if a car door is in the danger zone, personnel who have not left the car will be guided or warned by emergency stations and in-car broadcasts, guiding them to evacuate from the current car to other cars and then to the emergency station platform. The following algorithm is used:

[0135] The area outside the carriage is designated as a no-entry zone, and people are no longer evacuated through this area.

[0136] Instead of setting the car door as the starting point, the entrance to the safety car connected to it is set as the new starting point. At the same time, determine whether to temporarily block the car doors adjacent to the danger zone to prevent people from accidentally entering the danger zone. If blocking is necessary, remove them from the traversable area. Simultaneously, modify the h(n) values ​​for the danger zone and the target point.

[0137] Recalculate the optimal evacuation path from the new starting point (the entrance to the internal passage of the adjacent safety compartment) through the internal connecting passage of the compartment to the entrance of each horizontal passage.

[0138] In this embodiment, based on the real-time optimal evacuation path calculated by the above algorithm, the central system 10 sends an evacuation instruction feedback signal to all subsystems. Upon receiving the signal, the direction indicator arrows in the alarm module 2023 turn on the corresponding indicator lights, allowing evacuees to safely evacuate according to the direction of the indicator lights. If the evacuation environment changes, for example, if the area ahead of the evacuation path is classified as a hazardous area due to smoke deposition, the optimal evacuation path is updated in real time, and the feedback signal is also updated to each subsystem, changing the direction of the evacuation indicator lights to prevent people from entering the hazardous area and causing further casualties.

[0139] This embodiment can capture early fire signals in real time, collect and analyze disaster-causing factors within emergency rescue stations in real time, issue alarm signals within dangerous areas, and, based on the impact of disaster-causing factors in railway tunnel fires on personnel evacuation behavior, plan optimal evacuation routes in real time and direct personnel evacuation directions in fire situations. This embodiment not only protects the lives of passengers and staff and reduces property losses in the event of a fire accident, but also improves the safety management level of railway tunnels and ensures the safety of railway transportation. The implementation and application of this embodiment is expected to significantly enhance the safety assurance capabilities of railway tunnels and is of great significance to promoting the sustainable and healthy development of railway transportation.

[0140] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A fire environment emergency evacuation system for a railway tunnel emergency rescue station, characterized in that: include: Subsystems and a central system; the subsystems are installed at certain intervals on the side walls of the emergency rescue station, and the data collected by the subsystems are transmitted to the central system via serial communication. The central system is installed in the chambers at both ends of the tunnel emergency rescue station for storing emergency control cabinets; wherein: The subsystem includes: Data collection module, real-time monitoring of various disaster-causing factors in the area where the subsystem is located; The data analysis module uses a parallel computing algorithm with time filtering based on various disaster-causing factor data to determine the danger level of the area through threshold comparison. An alarm module, which issues an alarm according to the degree of danger and indicates the safe evacuation direction; The central system aggregates data from all subsystems, dynamically divides safe evacuation areas on a preset two-dimensional grid map, and calculates the optimal evacuation path from each grid to the safe area based on an improved D* algorithm tailored to the specific environment of railway tunnel fires, thereby avoiding dangerous areas. Ultimately, the optimal evacuation path is determined based on real-time changes in the fire environment, and the signal is then transmitted back to each subsystem to indicate the evacuation direction. The central system includes a data receiving and preprocessing module, a data storage module, and an optimal evacuation path analysis module; wherein: The data receiving and preprocessing module is used to receive raw data from various subsystems or external data sources, and perform preliminary cleaning, formatting, and standardization on the data to ensure data consistency and availability. After processing, the worker thread writes the preprocessed data into a temporary buffer. The main thread periodically extracts data from the temporary buffer and writes it in batches into the data storage module. A data storage module, used to store the pre-processed data in an appropriate storage system so that the optimal evacuation path analysis module can access it in real time and efficiently; The optimal evacuation path analysis module is used to build a dynamic two-dimensional grid map based on the real-time changing parameters of the tunnel evacuation environment. It then calculates the shortest path from each grid to the safe area based on the improved D* algorithm, and the optimal safe evacuation path for people in different areas. The signal is then transmitted back to each subsystem based on the optimal evacuation path. The optimal evacuation path analysis module is specifically used for: When the carriage door is in the safe area and there is no danger zone blocking the route to the cross aisle entrance, then: The optimal evacuation path is calculated using the A* algorithm, and the h(n) and g(n) values ​​for each grid cell are recorded. The h(n) value represents the heuristic function-estimated distance from the grid cell to the target point, which is the entrance to each cross-channel. The g(n) value represents the sum of the actual movement costs from the starting point to the grid cell. The h(n) value for each grid cell is initialized to the Manhattan distance to each cross-channel entrance. g(n) is initialized to infinity, indicating that no exploration has taken place. All cross-channel entrances are considered as candidate destinations. Specifically: Use the A* algorithm to perform a breadth-first search from the starting point to calculate the shortest path from the starting point to each entrance; Every time a grid is explored, its g(n) value is updated to the sum of the actual costs from the starting point to this grid; Record all optimal evacuation route plans; When the state of a certain grid is detected to be dangerous, its h(n) value is set to infinity, that is, the grid will not be included in the optimal evacuation path; With this grid as the center, add the 8 adjacent grids in the direction to the open list; Take a grid from the open list for modification, calculate the estimated cost g(n)+h(n) of walking from its adjacent grid to this grid, and compare it with the current g(n) value; If the new value is better, update g(n) with the new value and add the adjacent grid to the open list; Repeat the above steps until the open list is empty, which indicates the global optimal evacuation path. Finally, output the revised optimal evacuation path.

2. The fire environment emergency evacuation system of the railway tunnel emergency rescue station according to claim 1, characterized in that: The data acquisition module includes a temperature sensor, an oxygen concentration sensor, a thermal radiation sensor, a CO concentration sensor, and a visibility sensor; each sensor transmits the collected data to the data analysis module through analog signal transmission and digital signal transmission.

3. The fire environment emergency evacuation system of the railway tunnel emergency rescue station according to claim 2, characterized in that: The data analysis module includes a sensor interface, a microprocessor unit and a communication interface; wherein: The sensor interface is used to receive different types of environmental parameters transmitted by various sensors; The microprocessor unit is configured to receive the environmental parameter data transmitted by the sensor interface and preprocess the environmental parameters to convert them into a unified format that can be processed by the microprocessor unit; wherein, during the preprocessing, a timer is defined to match and align the environmental parameters of different sensors according to timestamps at a preset standard time interval to ensure data consistency and accuracy, and then the environmental parameters and corresponding timestamps are respectively filled into corresponding structure fields to obtain the environmental parameters in a unified format; The microprocessor unit is also used to compare the environmental parameters processed by different sensors with preset thresholds to determine whether the environmental parameters exceed their corresponding thresholds. Based on the situation of exceeding the thresholds, a time-filtered parallel computing environmental parameter safety judgment algorithm is used to make a safety judgment on the environmental parameters. When the area where the judgment subsystem is located is determined to be a dangerous area, an alarm signal is sent to the alarm module through the communication interface.

4. The fire environment emergency evacuation system for the railway tunnel emergency rescue station according to claim 3, characterized in that: The specific judgment process of the time-filtered parallel computing environment parameter security judgment algorithm is as follows: Define the threshold judgment function Fi (i=1, 2, ..., N) of N environmental parameters; The data Pi (i=1, 2, ..., N) collected in real time from N environmental parameters are simultaneously used as input and passed into each judgment function Fi for calculation; The judgment function Fi calculates whether the corresponding data Pi exceeds its corresponding threshold Ti in a parallel computing environment and records the result as Ri; where Ri is 0 or 1, 0 means that the threshold is not exceeded, and 1 means that the threshold is exceeded; Perform time filtering on the N results Ri; If Ri is 1 for consecutive ΔT time slices, the environmental parameter is ultimately judged to be above the threshold and the area is considered a dangerous area; otherwise, it is a safe area.

5. The fire environment emergency evacuation system for the railway tunnel emergency rescue station according to claim 1, characterized in that: The subsystem also includes a test / reset button, a backup power supply, and a status indicator; wherein the test / reset button provides a physical button for testing the alarm function or resetting the alarm status, and the status indicator uses an LED indicator light to display the working status of the module; the backup power supply is used to ensure that the subsystem itself has a stable power supply.

6. The fire environment emergency evacuation system for the railway tunnel emergency rescue station according to claim 1, characterized in that: The central system is the part that connects the subsystems and performs data processing, evacuation path analysis, and sends feedback signals. The central system uniformly processes the collected data and dynamically divides the safe evacuation area on a preset two-dimensional grid map. Based on the improved D* algorithm combined with the special environment of railway tunnel fires, it calculates the optimal evacuation path from each grid to the safe area to avoid the dangerous area. It can obtain the real-time optimal evacuation path based on changes in the evacuation environment, and finally transmits the signal back to each subsystem, indicating the safe evacuation direction based on the direction of the evacuation path.

7. The fire environment emergency evacuation system for the railway tunnel emergency rescue station according to claim 1, characterized in that: The optimal evacuation path analysis module is also used to: after determining the danger level of the area where the car door is located, when a car door is in the danger zone, the optimal evacuation path is generated for the people who have left the car using the following algorithm: Get the boundary coordinate point set of the danger zone; A boundary line is constructed based on all boundary coordinate points in the boundary coordinate point set, and the distance from the starting point to each point on this boundary line is calculated. The point with the minimum distance is selected as the temporary target point candidate on the boundary line. Select the candidate temporary target point with the smallest distance as the global temporary target point, and modify the h(n) values ​​of the dangerous area and the global temporary target point; Use the D* algorithm to calculate the shortest path out of the danger zone, then restore the h(n) value, restart the D* algorithm, and calculate the complete optimal evacuation path.

8. The fire environment emergency evacuation system for the railway tunnel emergency rescue station according to claim 7, characterized in that: The optimal evacuation path analysis module is also used to: After determining the danger level of the area where the car door is located, if a car door is in the danger zone, personnel who have not left the car are guided or warned by emergency stations and in-car broadcasts to guide people from the current car to other cars and then to the emergency station. The following algorithm is used to generate the optimal evacuation path: The area outside the carriage is designated as a no-traffic zone, and people are no longer evacuated through this area; Instead of setting the car door as the starting point, the entrance to the internal passage of the safety car connected to the car is set as the new starting point. At the same time, determine whether the car door adjacent to the danger zone needs to be temporarily blocked to prevent people from accidentally entering the danger zone. If blocking is required, remove it from the passable area and modify the h(n) values ​​of the danger zone and the target point. Recalculate the optimal evacuation path from the new starting point, that is, the entrance of the internal passage of the adjacent safety compartment through the internal connecting passage of the compartment to the entrance of each horizontal passage.

Images